Allelic-dependent expression of an activating Fc receptor on B cells enhances humoral immune responses

Abstract

B cells are pivotal regulators of acquired immune responses, and recent work in both experimental murine models and humans has demonstrated that subtle changes in the regulation of B cell function can substantially alter immunological responses. The balance of negative and positive signals in maintaining an appropriate B cell activation threshold is critical in B lymphocyte immune tolerance and autoreactivity. FcγRIIb (CD32B), the only recognized Fcγ receptor on B cells, provides immunoglobulin G (IgG)-mediated negative modulation through a tyrosine-based inhibition motif, which down-regulates B cell receptor-initiated signaling. These properties make FcγRIIb a promising target for antibody-based therapy. We report the discovery of allele-dependent expression of the activating FcγRIIc on B cells. Identical to FcγRIIb in the extracellular domain, FcγRIIc has a tyrosine-based activation motif in its cytoplasmic domain. In both human B cells and B cells from mice transgenic for human FcγRIIc, FcγRIIc expression counterbalances the negative feedback of FcγRIIb and enhances humoral responses to immunization in mice and to BioThrax vaccination in a human anthrax vaccine trial. Moreover, the FCGR2C-ORF allele is associated with the risk of development of autoimmunity in humans. FcγRIIc expression on B cells challenges the prevailing paradigm of unidirectional negative feedback by IgG immune complexes via the inhibitory FcγRIIb, is a previously unrecognized determinant in human antibody/autoantibody responses, and opens the opportunity for more precise personalized use of B cell-targeted antibody-based therapy.

Fig. 1. Expression of FcγRIIc protein in B cells

(A) Chromatograms showing the rs10917661 (nt202 T>C) polymorphism (amino acid position 13) in the first extracellular domain of FCGR2C. (B) RT-PCR detection of FCGR2A, 2B and 2C mRNA in EBV B cells containing either the FCGR2C-ORF (lane 1 and 2 from left) or -STP allele (lane 3). RT-PCR analysis of human myeloid cell line U937 cells using the same primers is used as a control for FCGR2A/B/C message (lane 4). Data are representative of 2 independent experiments. (C) Lysates of EBV B cells or primary human CD19⁺ cells were immunoprecipitated with mAb4F5 recognizing the FcγRIIb/c EC domain and then blotted with either anti-FcγRIIc CY pAb or anti-FcγRIIb CY pAb. FcγRIIc protein was detected as a protein with the FcγRIIb EC domain and FcγRIIc CY tail only in ORF B cells but not STP B cells. Data are representative of 3 independent experiments.

Fig. 2. Surface expression of FcγRIIc protein on human B cells

(A–D) EBV transformed human B cells from individuals containing the FCGR2C-ORF allele (A,B) or that are homozygous for the FCGR2C-STP allele (C,D) were opsonized with mAb4F5 (green) either alone or with crosslinking using a secondary antibody, then intracellularly stained for the FcγRIIc CY domain (red). (E) FCGR2C-ORF allele positive EBV transformed human B cells were incubated with a mAb specific to FcγRIIa EC and showed no staining, comparing to the positive staining of FcγRIIa EC in U937 cells. Results are representative of 3 independent experiments (Scale bar: 10μm). (F) Quantification of the EC (green) and the CY (red) staining of FcγRIIc or FcγRIIb colocalization after crosslinking, calculated by Metamorph software from 6–8 randomly chosen cells.

Fig. 3. Higher total FcγRII expression on B cells with increasing numbers of the FCGR2C-ORF allele

(A) Representative results of mAb 4F5 staining on EBV B cell lines derived from donors with different numbers of FCGR2C-ORF alleles. (B) A summary of increasing mAb 4F5 staining on EBV B cell lines derived from donors with different number of FCGR2C-ORF alleles (p = 0.017, ANOVA). (C,D) Summary of increasing mAb 4F5 staining of total FcγRII (FcγRIIb+FcγRIIc) expression on primary CD19⁺ B cells (C) and CD19⁺CD27⁺ memory B cells (D) from donors with different number of FCGR2C-ORF alleles (p = 0.042 and 0.0005 respectively, ANOVA).

Fig. 4. Activating properties of FcγRIIc in transduced A20IIA1.6 cells and primary human B cells

(A,B) Cellular tyrosine phosphorylation in FcγRIIc (A) or FcγRIIb (B) transduced IIA1.6 cells upon coligation with BCR. Cells were stimulated with equal molar amount of intact (25 μg/ml) or F(ab′)₂ fragment (16.6 μg/ml) of goat anti-mouse IgG for indicated time, affording BCR/FcγRIIc coligation or BCR crosslinking alone. Whole cell lysates were also re-probed for FcγRIIc (A) or FcγRIIb (B) to verify receptor expression and comparable protein loading. (C,D) Tyrosine phosphorylation of FcγRIIc (C) or FcγRIIb (D) upon co-crosslinking with BCR. (E and F) Tyrosine phosphorylation of Syk and BLNK detected by immunoprecipitation. Normalized intensities of both pSyk and pBLNK are significantly increased with FcγRIIc/BCR colligation, and decreased with FcγRIIb/BCR colligation comparing to BCR ligation alone (*E, P=0.002 (pSyk), 0.003 (pBLNK); F, P=0.033 (pSyk), 0.0097(pBLNK), ANOVA). Data are representative of 2 replicate experiments. (G) Ca²⁺ flux in different transductants evaluated by flow cytometry as Fluo-4/SNARF-1 ratio. Arrows indicate time points of adding indicated stimuli. Data are representative of 3 independent experiments. (H) Human PBMC were stimulated as indicated and the level of Syk phosphorylation was quantitated in CD20⁺ B cells by FACS. Grey line indicates baseline of pSyk in unstimulated cells. Net effect of FcγRIIc is analysed by comparing MFI of pSyk with equal molar amount of intact (15 μg/ml) vs F(ab′)₂ (10 μg/ml) of Goat anti Human-(IgG+IgM) stimulation (mean± s.e.m., n=6, P=0.024, t-test). (I) Ca²⁺ flux in primary B cells evaluated by flow cytometry. Data are representative of 5 independent experiments (P=0.032, t-test).

Fig. 5. Generation and characterization of B-cell specific FcγRIIc transgenic mice

(A) Schematic illustration of the fragment used to inject fertilized mouse eggs. (B) FACS analysis of FcγRIIc transgene and endogenous mouse FcγRIIb expression in splenic B cells from TG and NTG littermates. Data are representative of 3 independent experiments. (C) Demonstration of FcγRIIc protein in CD43⁻ splenocytes (CD43⁻=B cells, CD43⁺= non-B cells) by western blot. CD20 and β-actin blots are shown as controls. (D) Tyrosine phosphorylation of FcγRIIc after co-engagement with intact anti-Ig. Data are representative of 2 independent experiments. (E) Syk phosphorylation in splenic B cells stimulated as indicated. Relative pSyk is analysed by normalizing intensity of pSyk bands to corresponding Syk bands, then compare intact (15 μg/ml) vs F(ab′)₂ (10 μg/ml) goat anti–mouse IgM stimulation (mean± s.e.m. n=2, P=0.015, t-test). (F) Ca²⁺ flux in splenic B220⁺ cells was evaluated by flow cytometry as Fluo-4/SNARF-1 ratio. Arrows indicate time points of adding indicated stimuli. Data are representative of 3 independent experiments. (G) Surface BR3 expression on splenic B cells after 72 hours of treatment with indicated stimuli. Net effect of FcγRIIc is analysed by comparing MFI of BR3 with equal molar amount of intact (15 μg/ml) vs F(ab′)₂ (10 μg/ml) goat anti–mouse IgM stimulation (mean± s.e.m. n=3, P=0.024, t-test).

Fig. 6. Enhanced humoral immune responses in B-cell specific FcγRIIc TG mice

8–12 week old male TG mice and their NTG littermates were intraperitoneally injected with either 25ug TNP-Ficoll (A,B, n=10 and 7 for TG and NTG respectively) or 50ug TNP-CGG/alum (C,D, n=12 and 9 for TG and NTG respectively) on day 0 and bled at various time points as indicated. For T-dependent experiments, animals received a secondary immunization on day 28. Serum antibody titers were measured by ELISA using TNP₂₆-coated plates. Data are presented as mean ± s.e.m. and group effects were analyzed by repeated measure ANOVA. (E) Production of high affinity antibody after boost was measured by ELISA using TNP₂-coated plates (P=0.0097 for D40 and 0.027 for D51, t-test).

Fig. 7. Impact of FCGR2C-ORF in human vaccine response and susceptibility to autoimmunity

(A) Levels of Ab specific for the Anthrax protective antigen (AbPA) in vaccinated donors homozygous for either FCGR2C ORF or STP allele were assessed at the indicated time points. AbPA levels determined by ELISA were analyzed by ANCOVA and AbPA levels at the earliest time points in the vaccine study are higher in donors homozygous for the ORF allele with significantly significance reached at the 4 week time point (p<0.02). (B,C) The cumulative proportion of FCGR2C-ORF positive participants with increasing gene copy number in (B) African Americans (AA), and in (C) European Americans (EA) in patients with SLE and healthy controls. (D) Determination of the effect size of the FCGR2C-ORF allelic association with the risk of development of SLE. The odds ratio (OR) and p values, determined by logistic regression, between the number of FCGR2C-ORF alleles and the lupus phenotype are shown.